Results for egcs-2.91.52 19980727 (gcc2 ss-980609 experiment

Joe Buck jbuck@synopsys.com
Thu Aug 6 12:44:00 GMT 1998


> | 	If you want correct IEEE operation from IBM POWER architecture (or
> | common-mode within AIX), one must use the GCC -ffloat-store option to
> | prevent the excess precision.  AIX XLC has a similar option.  Maybe this
> | option should be used for all testcases in the execute/ieee subdirectory.
> 
> If adding -ffloat-store to the test cases solves the problem that it
> FAILs, then that seems a good solution to me :).
> Maybe you can write a patch and post it to cvs-patches for more discussion?
> I'd do it, but I don't have this architecture to test anything.

It would seem that any test of IEEE conformance that will fail if the
processor uses extra precision must specify -ffloat-store to be a valid
test.  The patch itself is trivial, so I suppose the real work is figuring
out which test cases should be patched.






More information about the Gcc mailing list